J. Victor Owen

An empirical sapphirine-spinel MgFe exchange thermometer and its application to high grade xenoliths in the Popes Harbour dyke, Nova Scotia, Canada

Owen, J.V. and Greenough, J.D., 1991. An empirical sapphirine-spinel exchange thermometer and its application to high grade xenoliths in the Popes Harbour dyke, Nova Scotia, Canada. Lithos, 26: 317-332. Spinel-sapphirine-corundum-rutile parageneses in metapelitic xenoliths from the lamprophyric Popes Harbour dyke are enclosed by feldspathic ( _+ rare quartz ) haloes that embay aluminosilicates and biotite. These feldspathic haloes contain plagioclase (An2o-4o) and/or an alkali or ternary (hypersolvus) feldspar, and show a variety of igneous and devitrification textures, suggesting an anatectic origin. The spinelbearing parageneses are interpreted as the refractory residue formed by the incongruent melting of biotite, aluminosilicates and associated phases. Equilibration temperatures of these assemblages are estimated from an empirical sapphirine-spinel Mg-Fe exchange thermometer derived from literature data on both silica-saturated and undersaturated sapphirine granulites. Linear regression (R=0.81) of the calibrant data yields the expression T(°C) = [800+ (228*lnKd) ] -273 where Kd= ( X~/ X~g) / ( X~/ X~g) Precision is estimated at _+ 100°C, but will likely be less for highly oxidized sapphirine granulites owing to ( 1 ) errors in the stoichiometric estimation of XFe3+ from microprobe data and (2) the formation of magnetite at the expense of spinel or sapphirine, leading to an increase in XMg in either or both phases during cooling. Application of this expression to the reduced (graphite-bearing) Popes Harbour xenoliths yields T of 725-795°C. Anatexis is attributed to thermal metamorphism by the lamprophyric magma prior to and/or during entrainment of xenolith material in the dyke. High-T assemblages were quenched before the xenoliths attained thermal equilibrium with the magma, consequently prograde reaction textures and compositional zoning patterns are preserved.

Platinum-group element geochemistry of continental tholeiites: Analysis of the Long Range dyke swarm, Newfoundland, Canada

Abstract The rift-related, Late Proterozoic Long Range continental dykes of western Newfoundland have platinum-group element (PGE; Pd, Pt, Rh, Ru and Ir determined) concentrations that correlate with indicators of magma evolution [e.g., Mg ′ = Mg ( Mg + 0.9 Fe tot ) ] and magma alkalinity (La/Sm), suggesting that metamorphism had a minimal effect on the PGE. The low-melting-point (LMP) PGE (Pd, Pt and Rh) and Cu behaved incompatibly as magma evolved, indicating that sulphide fractionation did not occur and that the dykes hold little economic promise. Hypotheses that augite (a major fractionating phase) has an affinity for the PGE are not supported. Primitive rocks tend to have lower Ir than evolved rocks and all samples display a negative high-melting-point (HMP) PGE (Ru and Ir) anomaly on mantle-normalized diagrams. Olivine and early chromite fractionation may have decreased the concentrations of these elements. The Long Range dykes confirm that Pd/Ir ratios and absolute LMP PGE concentrations tend to decrease with increasing hot-spot magma alkalinity. The lower PGE concentrations in most MORB compared with hot-spot basalts imply that there are at least two mantle reservoirs for the PGE. Lithosphere overlying mantle sources (continental or oceanic) has little effect on PGE concentrations. The apparently similar source-region PGE concentrations for hot-spot basalts suggest that differences in PGE concentrations between MORB and hot-spot basalts reflect incomplete segregation of the PGE into the core during its formation and rise of the homogeneously dispersed elements to upper-mantle depths in deep mantle plumes.

Provenance of temper in a New Kingdom Egyptian pottery sherd: Evidence from the petrology and mineralogy of basalt fragments

A New Kingdom spinning bowl from Karnak (Luxor) Egypt is similar in form to spinning bowls commonly found at other Egyptian sites and has a bulk chemical composition in the range for other Egyptian marl vessels. These data support a domestic origin. The matrix of the bowl contains unaltered, sand-sized, mafic rock fragments with volcanic, subophitic textures. Over 20% of the sand-sized grains consist of angular, unweathered rock fragments, and of these ∼20% are volcanic. Apparently they were added as temper. Electron microprobe analyses show that augite, plagioclase, and, where present, pigeonite, in nine of these have compositions typical of mafic igneous rocks. Geothermometry confirms crystallization at ∼1100°C. Pyroxene discrimination diagrams indicate geological sources ranging from within-plate alkali basalts to within-plate, continental tholeiites. Suitable sources for the temper are rare in Egypt. Both alkaline and tholeiitic, postorogenic (unaltered) late Cenozoic basalts occur in the Cairo area, making this the most likely but not the only possible source for the temper. The pottery may have also originated in Cairo because raw rock materials were moved upstream less commonly than down the Nile. A Cairo provenance for the Karnak artifact is consistent with the everyday movement of people and goods along the Nile between the ancient twin capitals of Memphis and Thebes. These results and the common occurrence of volcanic rocks as temper indicate that microbeam analytical techniques may help narrow the provenance of ancient pottery.

Metamorphic geology and regional geothermobarometry of a Grenvillian massif: the Long Range Inlier, Newfoundland

Abstract Geothermobarometry of Middle Proterozoic, amphibolite to granulite facies gneiss of the Long Range Inlier reveals a southwestward increase in metamorphic temperature from 600–650°C to nearly 800°C, at a pressure between 5 and 8 kbar throughout the massif. A steep d P d T slope of 14 bars °C−1 determined using the garnet-orthopyroxene (-plagioclase-quartz) thermobarometer indicates fairly rapid uplift of the inlier and (or) records the effects of a late Grenvillian (∼970 Ma) tectonothermal overprint. Grenvillian granitoid rocks and a mafic pluton of unknown age have narrow thermal aureoles, and intruded the gneiss complex following regional, high-grade metamorphism. Metamorphic conditions of the adjacent Grenville Province span those determined for the Long Range Inlier. Available data suggest that both areas were uplifted from moderate crustal levels during the Middle to Late Proterozoic, and indicate negligible differential uplift of the inlier relative to the Grenville Province during Paleozoic orogenies.

Sagged phosphatic Nantgarw porcelain (ca. 1813–1820): Casualty of overfiring or a fertile paste?

Nantgarw phosphatic (bone-ash) porcelain (ca. 1813–1820) is renowned for its translucency and the high quality of its enamel decoration. However, only a small proportion of its wares (perhaps 10%) were successfully fired due to sagging (body distortion) and other problems. This indicates that: (1) In terms of its potential for generating a minimum melt, the Nantgarw paste had an unusually fertile composition, or (2) Nantgarws staff had difficulties in controlling kiln temperatures, so that its wares tended to be overfired. This issue has been addressed using analytical data for sherds excavated from the factory site. Detailed modal, petrographic, and geochemical data demonstrate that extensively sagged wasters contain a high proportion (∼43–63 vol %) of a former melt phase but are surprisingly porous (7–25 vol % pores). In terms of minimum melt generation, the sagged samples have a more fertile bulk composition than their successfully fired counterparts. Most of the wasters contain subsolidus anorthite (∼An95) enclosed by a melt phase with (once corrected for entrained silica polymorph crystallites) a eutectic (minimum melt) composition. The anorthite, however, was completely resorbed by the melt in some of the sagged samples, which subsequently crystallized liquidus anorthite that displays a quenched morphology. These samples were therefore fired above (Tmax <1430°C, as estimated from melt compositions) the eutectic (T∼1290°C) in the anorthite–tricalcium phosphate–silica system. The fact that other wasters contain subsolidus anorthite and a minimum melt shows that firing at the eutectic did not guarantee a successful firing, regardless of the melt fertility of the paste. It is likely that the duration of firing near Tmax and character of the object (i.e., flatware vs. hollow-ware) were also important variables in this regard. In addition to producing their well-known phosphatic wares, Nantgarws proprietors apparently also experimented with silicious pastes with compositions akin to true porcelain. Some of silicious sherds found at the site have a lead-bearing (∼15 wt % PbO) glaze, indicating that they survived an early, high temperature biscuit firing, after which they were fired at lower temperature in the glost kiln (i.e., a “soft-paste type” firing sequence). Evidence that the Nantgarw kiln could achieve temperatures in the order of 1400°C further suggests that these silicious sherds are b.f. wasters.

Magma injection directions inferred from a fabric study of the Popes Harbour dike, eastern shore, Nova Scotia, Canada

A 15-m-wide dike at Popes Harbour on the eastern shore of Nova Scotia shows an increase in both the size (up to 1 m) and density (up to ∼50%) of predominantly pelitic xenoliths toward the dike center. These features reflect flowage differentiation processes during dike emplacement. Flow direction is indicated by sillimanite xenocrysts that represent the unassimilated products of xenolith disaggregation. The xenocrysts are highly aligned parallel to the dike margins (northwest-southeast), with a mean plunge of 20°SE, suggesting that the magma last moved, and may have been injected, in a subhorizontal, northwestward direction. The available information indicates that lateral injection of magma in mafic dikes is more common than predominantly vertical movement. This study shows that an analysis of fabrics in mafic dikes should be useful in further evaluating the role of lateral magma injection in dike formation.

Syn-metamorphic element transfer across lithological boundaries in the Port-aux-Basques gneiss complex, Newfoundland

Mineralogical reaction-zones developed between mafic gneiss (amphibolite) and metapelite reveal the role of cm-scale metasomatism during amphibolite-facies metamorphism of the Port-aux-Basques gneiss complex (PBGC). Ionic diffusion between mafic and pelitic layers led to the development of 1–3 cm wide, schistose, biotite + garnet-rich (Type 1) reaction-zones at the margins of mafic layers, and/or the crystallization of poikiloblastic hornblende within a garnet- and biotite-depleted (Type 2) reaction-zone up to 20 cm wide within nearby paragneiss. Garnet-biotite thermometry of the Type 1 reaction-zones indicates Tmax of c. 560–645°C at a pressure of c. 6 kbar constrained by “GASP” and “GRAIL” subassemblages in the host rock. Compared to the “unaltered” amphibolite, Type 1 reaction-zones are enriched in K, Rb and Ba, and depleted in Ca and Sr; compared to the “unaltered” metapelite, Type 2 reaction-zones show opposite trends: they are depleted in K, Rb and Ba, and enriched in Ca and Sr. This indicates that the formation of the reaction-zones involved the exchange of K, Ca and related trace elements in opposite directions across the amphibolite/paragneiss interface, and that the system was approximately closed to these components where both reaction-zones are present.

Nicholas Crisp's “Porcellien”: A petrological comparison of sherds from the Vauxhall (London; ca. 1751–1764) and Indeo Pottery (Bovey Tracey, Devonshire; ca. 1767–1774) factory sites

The character of porcelain wares made by Nicholas Crisp early and late in his career was assessed using microchemical and petrographic data for sherds excavated from the sites of the factories he operated at Vauxhall and Bovey Tracey. The results indicate that, over time, Crisp increasingly made use of diverse types of pastes as he struggled to produce a commercially viable line of porcelain. Based on the analysis of a limited number of samples, he appears to have largely restricted himself at Vauxhall to using soapstone (Mg-rich)- and flint-glass (Pb-rich) frit-bearing pastes that varied in the amount of calcite they contained. He also experimented with Mg+Pb-rich pastes at Bovey Tracey, but included a novel ingredient (barite) and varied the proportion of other minor constituents (e.g., bone ash), apparently in an effort to resolve some of the firing problems that plagued him at Vauxhall. In addition, Crisp appears to have produced bone ash (phosphatic) porcelain at Bovey Tracey, and, in collaboration with William Cookworthy, the proprietor of the Plymouth factory, fired a range of true porcelain (Si+Al-rich) pastes. Bulk compositional data indicate that Crisps diopside-bearing Mg+Pb-rich wares were derived from pastes containing talc and calcite rather than dolomite. The mineralogy of these and some contemporary magnesian/plombian porcelains are interpreted using the SiO2-CaO-MgO phase diagram. This diagram shows that these wares can form and preserve diopside (Ca-Mg silicate) given suitable bulk CaO contents and kiln-firing temperatures. Phosphatic sherds from Bovey Tracey are compositionally distinct (lower SiO2 and higher Al2O3 and bone-ash components) from a single bone-ash sample from Vauxhall, indicating that Crisp experimented with novel bone-ash pastes, and was not positively influenced by the Vauxhall phosphatic recipe, if indeed one existed. True porcelains from Bovey Tracey have more extreme SiO2/Al2O3 ratios (= 2.0 [two sherds]; 4.5 [one sherd]) than their Plymouth/Bristol counterparts (SiO2/Al2O3 = 2.3–3.0). Collectively, the analytical data underscore the experimental—and ultimately unsuccessful—character of the diverse wares produced by Nicholas Crisp.

Provenience of eighteenth-century british porcelain sherds from sites 3b and 4e, fortress of Louisbourg, Nova Scotia: Constraints from mineralogy, bulk paste, and glaze compositions

Twenty-six British porcelain sherds excavated from two properties at the Fortress of Louisbourg were analysed by electron microprobe with the objective of identifying the factories from which they originated (that is, their provenience). Nineteen of the samples are phosphatic; seven are magnesian. Sixteen of the phosphatic sherds have paste and glaze compositions consistent with Bow porcelain (ca. 1747 to 1776). This attribution is supported by underglaze blue patterns (Dragon, Cannonball, and Desirable Residence patterns) and famille rose overglaze polychrome designs that match the decoration used on Bow porcelain. A calcic plagioclase-bearing sherd has a composition suggesting derivation from the Gilbody works (Liverpool, ca. 1755 to 1761). Another unassigned sherd has a paste composition that resembles products of the Lowestoft factory (ca. 1757 to 1799), but contains plagioclase, a mineral not known to occur in Lowestoft porcelain, and its glaze contains small amounts of tin, a component unknown in analysed Lowestoft glazes. One highly porous sample has an anomalous composition (a lower phosphate content than any known bone-ash porcelain), and appears to have been chemically modified in the ground. It, too, remains unattributed. The magnesian sherds contain diopside and enstatite, and thus are mineralogically similar to another type of Liverpool porcelain (late Chaffers [ca. 1756 to 1765]) and its successor, the Christian/Seth Pennington works; ca. 1765 to 1799). Compositionally, the body and glaze of these samples resemble Chaffers/Christian/Seth Pennington porcelain. This attribution is further supported by their underglaze blue patterns, which match some of those (Liver Bird pattern) known to have been used at the Liverpool works. Contemporary documents record the fact that the Bow works exported significant amounts of its wares to North America. The discovery of Chaffers/Christian/Seth Pennington porcelain at Louisbourg and other archaeological sites in Atlantic Canada (Fort Beauséjour) indicates that some Liverpool factories also supplied colonial markets with porcelain.

Compositional constraints on the identification of Eighteenth-Century porcelain sherds from fort Beauséjour, New Brunswick, and Grassy Island, Nova Scotia, Canada

Seven samples of 18th-century British soft-paste porcelain from Fort Beauséjour and Grassy Island were analyzed by microchemical techniques. Four of the samples have Mg+Pb-rich compositions; three are phosphatic. Two of the Mg+Pb-rich samples are attributed to the Dr. Wall period (ca. 1751–1776) of the Worcester manufactory. The other two, however, contain diopside, a mineral not known to occur in Worcester porcelain. One of the phosphatic samples contains sulphate, and compositionally resembles Bow porcelain; two sulphate-free phosphatic samples remain unattributed. Compositional constraints on the origin of these sherds has allowed for a tighter research focus, leading to pattern identification and more precise dating. Three of the Mg+Pb-rich sherds—including one diopside-bearing sample—apparently display the “cannonball” pattern used at Worcester between ca. 1756 and 1785. This feature, together with the Dr. Wall period attribution of the diopside-free samples, brackets their production to ca. 1756–1776. The diopside-bearing sherds likely were made by one of Worcester’s imitators, perhaps Caughley. The decoration on the Bow sherd suggests an early 1760s date. Where undisturbed, there is good correlation between archaeological context and the dated sherds.